Method of determining tolerance interval limits, method of evaluating a production process and corresponding computing device

Abstract

A method for determining tolerance interval (TI) limits is disclosed, comprising: providing a plurality of sample values ​​(x), wherein the sample values ​​(x) fluctuate and defining a sample value distribution (306, 602), the sample values ​​(x) being technical parameter values ​​associated with a sampling item of the sample, wherein the sampling item is a part of a drug delivery device (100), a component for a drug delivery device (100), or a drug delivery device (100), wherein the sampling items have the same construction, and wherein the technical parameter is limited by at least one technical limit value; selecting a probability distribution function (306, 602) based on the technical parameter and / or the sample values ​​(x); and determining a cutoff value (ξ) of the probability distribution function (306, 602) using the technical limit value (308). L ξ R ); the probability content (pc) specifying the tolerance interval (TI); and the limits (UTBL, LTBL) providing the tolerance interval (TI) for the technical parameters based on the transformation probability content, wherein the transformation probability content is based on a cutoff value (ξ). L ξ R ) and specified probability content (pc).

Description

[0001] This disclosure relates to a method for determining tolerance interval limits, a method for evaluating a production process, and a corresponding computing device.

[0002] ISO (International Organization for Standardization) 16269-6 Statistical Interpretation of Data - Part 6: Determination of Statistical Tolerance Intervals, 2014(E), or higher or lower versions, i.e., older or newer versions, relates to a method for estimating at least one limit of a tolerance interval that is valid for the production process of the apparatus.

[0003] There exist tolerance intervals that are restricted on both sides, i.e., bilateral tolerance intervals. However, tolerance intervals that are restricted on only one side and open on the other side can also be used, i.e., unilateral tolerance intervals. "Open" in this context may mean that no tolerance is specified in one direction for a certain technical metric, i.e., an upper or lower limit.

[0004] The tolerance interval calculation according to ISO 16269-6 applies to tolerance intervals for technical parameters with a distribution function, where, in principle, all values ​​may exist. However, some technical parameters may have technical limits, either an upper limit that cannot be exceeded or a lower limit below which the subsequent parameter will not take a value. For example, force may always be greater than zero.

[0005] The purpose of this disclosure is to provide an improved method for determining tolerance interval limits. The method preferably takes into account the influence of technical limits on the calculation of tolerance interval limits. Furthermore, an improved method for evaluating production processes is disclosed. Corresponding computer program products and computing devices for performing the method are also provided.

[0006] This objective is achieved by the method according to claim 1. Further embodiments are given in the dependent claims. Furthermore, this objective is achieved by a further method for determining tolerance interval limits, a method for evaluating a production process, a computer program product, and a computing device.

[0007] In one aspect, methods for determining the limits of tolerance intervals include:

[0008] -a) Provides multiple sample values, wherein the sample values ​​fluctuate and define a sample value distribution, the sample values ​​being technical parameter values ​​related to the sampled items, wherein the sampled items are parts of a drug delivery device, components for a drug delivery device, or drug delivery devices, wherein the sampled items have the same construction and are preferably manufactured according to the same specifications, and wherein the technical parameters are limited by at least one technical limit value.

[0009] -b) Select the probability distribution function based on technical parameters and / or sample values.

[0010] -c) Use the technical limit value to determine the cutoff value of the probability distribution function.

[0011] -d) Specifies the probability content of the tolerance interval, and

[0012] -e) Provide (preferably calculated) tolerance interval limits or at least two tolerance interval limits for technical parameters based on the transformation probability content, wherein the transformation probability content is based on a cutoff value and a specified probability content.

[0013] The order of steps a) through e) can be varied. However, steps a) through d) can be performed before step e), and step e) is based on the results of steps c) and d). Step c) is based on the results of step b). Therefore, step b) can be performed before step c). For example, the order of steps c) and d) can be interchanged.

[0014] This method can be used to determine the limits of a tolerance interval, wherein at least one limit of the tolerance interval can be used for technical purposes, particularly for the manufacture of drug delivery devices or related parts or components, and / or wherein statistical tolerances are used for manufacturing quality control and / or monitoring of drug delivery devices or related parts or components, particularly by means of:

[0015] -f) At least one limit of the tolerance range can be compared with the limit of the specification range of the production item, which has the same construction or the same construction design as the sampling item and is preferably manufactured according to the same specifications, such as product specifications or manufacturing specifications.

[0016] -g1) Based on the comparison results, if at least one limit of the tolerance interval is located within or within the production specification range, the production process of the production item can begin, continue, or restart.

[0017] -g2) Based on the comparison results, alternatively, if the at least one limit exceeds the specification range, the production process of the production project may not start or stop.

[0018] Steps a) through e) may be performed before step f). Step f) may be performed before step g1) or g2).

[0019] This method can be used to determine the limits of a tolerance range, wherein at least one limit of the tolerance range can be used to adjust the production process, particularly for the production process of manufacturing drug delivery devices or related parts or components.

[0020] The probabilistic content can be, for example, as defined in ISO 16269-6:2014(E), which will be explained in more detail below. Using the transformed probabilistic content is a simple way to consider the technical limit affecting the effective area of ​​the assumed probability distribution function. Unlike other methods that attempt to consider technical limits, it is not necessary to transform the probability function, as this can lead to questionable results. Furthermore, complex algorithms, especially iterative algorithms, are not necessary, and even exact integration is not required, both of which can consume significant computational power and time on digital processors. Moreover, the results of the proposed method are understandable and sufficient for practical applications, particularly in the field of medical devices.

[0021] The same method can also be applied to devices other than drug delivery devices, preferably to other medical devices or other industrial devices.

[0022] The term "identical construction" can refer to identical construction designs, such as having the same shape, the same mechanical properties, being based on the same CAD (computer-aided manufacturing) data, being manufactured in the same mold (injection mold), and / or being manufactured in the same cavity of the injection mold. Items with identical constructions can differ from each other solely due to manufacturing tolerances.

[0023] A component can be a single or separate element. At least two components functionally connected to each other can form an assembly. A drug delivery device can be a means of performing the motion required to dispense a drug from a reservoir containing the drug. A drug delivery device can be an injection device and / or may include a reservoir filled with the drug.

[0024] Specifications can be product specifications or manufacturing specifications. Specifications can specify upper and / or lower limits that must be followed.

[0025] A probability distribution function can be selected, which is or should be an indication of the distribution of sample values, preferably an indication of the distribution of basic sample values.

[0026] For a univariate (one-dimensional) probability distribution function, only one limit can be provided or calculated. For a multivariate (multidimensional) probability distribution function, if multiple technical limits are involved, multiple limits can be calculated according to the proposed method. An example of a multivariate distribution function is the bivariate distribution function.

[0027] Providing tolerance limits may include the ability to perform calculations. Alternatively or additionally, lookup tables may be used. Providing may also include calculations performed anywhere, for example, at different locations and / or on different computing devices, rather than on the device used to provide probabilistic content and / or transform probabilistic content and / or technical limit values.

[0028] If the industrial production process of the device is to be realized, it may in principle involve the following stages: design and manufacturing of one or more prototypes, small-batch production, medium-batch production, and large-batch production. It is recommended to begin the next stage only when tolerances and other specifications are clearly achievable or likely to be met within that stage. The proposed method allows for the calculation of tolerance range limits and can serve as the basis for further decisions on whether to begin the next stage or whether to take other measures (i.e., redesign or select a different or modified production method).

[0029] However, even after the production process has begun, the equipment parameters may be modified, which could affect the already evaluated parameters. Therefore, further evaluation may be necessary, and the proposed methods may be applied or reapplied.

[0030] Because the technical limits are considered when calculating one or two limits of the tolerance interval, the proposed procedure may be particularly advantageous for the alignment procedure.

[0031] Sample values ​​can also be referred to as observations. A sample can be a random sample from a large number of devices or device parts. Alternatively, the sample can include all currently available drug delivery devices, drug delivery device parts, or drug delivery device components (e.g., those produced in small batches in preparation for an industrial production process).

[0032] The limits of tolerance zones or update intervals can be calculated or estimated based on the statistical parameters of the sample values. The sample values ​​can be determined or measured, where the measurement includes consideration of SI (International System of Units) units.

[0033] The sample size can be less than the number of available or future-produced devices, for example, less than 10% or less than 1%, but greater than 0.001%. The envisioned production process can be, for example, a production process within a defined time period, such as a day, a week, a month, etc. The number of devices produced within a production cycle can be greater than 100, greater than 1,000, or greater than 10,000 devices. However, the number of devices produced may be less than 100 million devices.

[0034] The probability distribution function can be assumed to be normally distributed. It can be tested using, for example, the Anderson-Darling test or other appropriate tests. The estimated mean and estimated standard deviation can then be calculated and used in the proposed method. However, other probability distribution functions can also be used, such as the log-normal distribution, the Weibull distribution, the Gumbel distribution, or the Fréchet distribution. Other statistical descriptors, such as the estimated median, can also be used.

[0035] The specified probability content can refer to a proportion or percentage of the total number of production items, where the production items are parts, components, or devices of the same construction as the sampled items, such as drug delivery devices or other devices. This number of production items can be either already produced or can be produced in the future. This number can be at least 10, 100, or 1000 times larger than the sample size from which the sample values ​​were extracted.

[0036] According to ISO 16269-6:2014(E), the probability content is named proportionally, and its meaning is more easily understood if confidence values ​​are also considered. A confidence level of 1–α can be used, which specifies the probability that the tolerance interval for which the technical parameter must be estimated will contain at least the proportion of production items specified by the probability content. Alternatively, a determined confidence level can be used to calculate at least one limit of the tolerance interval. Conversely, α is used to estimate the probability that the tolerance interval contains production items less than the stated proportion.

[0037] Therefore, the distribution of sample values ​​may have a technical limit. A mathematical probability distribution function can simulate the distribution of sample values. To reflect the technical limit of sample values, it may be necessary to truncate the probability distribution function at a cutoff value representing the technical limit.

[0038] At least one or at least two descriptive parameters of the sample values ​​can be calculated and used to compute the cutoff value of the probability distribution function. The cutoff value can be determined by assuming a mapping or superposition of the probability distribution function on a histogram of the sample values. The following will combine... Figure 3 This will be explained in more detail. Statistical descriptive parameters can be the mean and standard deviation, especially when a normal distribution function is used in the calculation.

[0039] The physical parameters can be one of the following parameters of the drug delivery device:

[0040] a) Dosage accuracy, b) Picking torque, c) Dispensing force, d) Cap attachment force, e) Cap removal force, f) Needle cover removal force, g) Injection time, h) Activation force, i) Needle cap blocking distance, j) Needle extension, k) Discharge volume, or l) Assembly force. All of these parameters may be subject to technical limits. The proposed method allows for the calculation of reliable data on tolerance range limits while taking into account the technical limits of these parameters. However, tolerance ranges for other physical parameters, particularly thickness or length parameters, or at least their tolerance limits, can also be calculated using the proposed method.

[0041] a) Dosage accuracy: Dosage accuracy may relate to a device that only has a predetermined dose or a device that allows different doses to be selected. The injection device may be an autoinjector that injects a constant volume of fluid. Alternatively or additionally, the dose may be selected or chosen within a specific range. In the latter case, dosage accuracy is an indicator of the precision with which the selected dose corresponds to the expelled dose.

[0042] Doses below 0 ml (milliliters) are impossible; there is a technological limit. This may be particularly relevant to 1 IU testing. International units (IU) are related to the physiological effects of a drug / pharmaceutical on the human body. Dosage accuracy measures the amount of fluid expelled from the device, especially from a drug delivery / injection device. The drug may be insulin or another drug used to treat diabetes. Alternatively, the drug may be used for hormone therapy (thyroid) or other treatments. The drug may be one of the drugs listed below.

[0043] Dosage may be related to the amount of medication in the container, such as the cartridge, ampoule, or pre-filled syringe. The dose in the container may differ from the dose dispensed, see below (k).

[0044] b) Selecting Torque for Dosage Selection: Insulin pens or other drug delivery devices may include a dose / selection button (operating element) for selecting the dose. By rotating the button, the pen mechanism prepares to dispense the selected dose. A torque is applied to rotate the button; the corresponding torque can be measured or determined in testing. The torque is always above zero because no mechanism can reverse the selected dose. Optionally, the dose / selection button may be connected to a spring in the pen or device mechanism. In this case, the spring is loaded by rotating the button (operating element). A torque is applied to load the spring; the corresponding torque can be measured or determined in testing. Because loading the spring generates potential energy, the torque is always above zero. The selecting torque may not be the same if the doses selected are the same. However, it should be within the range defined according to the specification limits.

[0045] c) Dispensing force: A dispensing force may be required when dispensing medication manually or via an energy storage element (e.g., using a spring). This can be measured using a 10 N (Newton) counterweight. For example, the dispensing force may have to be generated by the user of the medication delivery device or by the spring. The user may be a child, the elderly, or someone with limited mobility or reduced dexterity.

[0046] d) Cap Attachment Force: The cap of the drug delivery device can be fastened to the device, for example, using a snap-on feature. The cap may cover the needle of the drug delivery device. The cap may need to be replaced after use to prevent injury or the spread of disease (if the needle of the drug delivery device is accidentally touched). The cap attachment force should not be too great so that even a weakened user, such as a user with limited mobility or reduced dexterity, can attach the cap. On the other hand, the cap must be properly secured to prevent accidental removal. The proposed method provides reliable cap attachment force values ​​that can be used to estimate whether the device is within its specification limits.

[0047] e) Cap Removal Force: The cap should not be fastened too loosely to prevent accidental removal and injury from accidental contact with the needle. On the other hand, the cap should not be fastened too tightly to allow for removal by physically weak users (e.g., those with limited mobility or reduced dexterity). The proposed method yields reliable cap removal force values ​​that can be used to estimate whether the device is within its specification limits.

[0048] f) Needle cap removal force: The cap, which contains the needle cap, can be removed from the autoinjector or another drug delivery device. When the cap is removed together with the needle cap, the autoinjector is ready for activation. Depending on the device design, the needle cap removal force may be the same as or different from the cap removal force.

[0049] g) Injection time: Again, this feature is particularly relevant to autoinjector devices. Users should be able to rely on a constant injection time.

[0050] h) Activation force: This can be the force that the user must exert to activate drug delivery, such as by pressing a button on the proximal end of the drug delivery device or moving another triggering component. This is an important parameter of the drug delivery device. The proposed method allows for reliable estimation of this parameter's value.

[0051] One type of activation force is the cap activation force, which may be associated with an autoinjector device. After the cap / needle cover is removed, the needle remains protected by the cap. This can be a push-back plastic part. This is done by pressing the autoinjector against the body. The force required to push the cap back is called the cap activation force. For some autoinjectors, an activation button may also need to be pressed. Therefore, identifying or measuring these two different types of activation forces can be useful.

[0052] i) Needle cap blocking distance: This is particularly relevant to autoinjector devices. The needle of an autoinjector can be protected by a movable needle cap before and after injection. The blocking force can be related to preventing proximal axial movement of the movable needle cap after injection. This protection can be tested by applying a force, such as 50 N (Newtons). The distance the needle cap moves can then be measured. This blocking prevents injury after using the drug delivery device. Reliable data on the tolerance limits for this parameter are particularly important.

[0053] The lower limit of the technology might be 0mm. More specifically, the distance might be 0mm when the force sensor detects the first contact with the needle cap. The upper limit of the technology could be defined by the total length of the needle cap extension. However, this is usually less relevant because the specification limits are much smaller.

[0054] Measuring the blocking distance can be one method of conducting the test. Measuring the blocking force can be an alternative method for testing the needle cap blocking force after injection.

[0055] j) Needle extension: Needle extension can be the distance the needle extends out of or from the autoinjector during injection. This extension can be a characteristic of the injection depth. Injection depth can depend on the length of the needle and / or its mounting position.

[0056] k) Discharge volume: The discharge volume must not exceed the filling volume of the container (e.g., cartridge / syringe). Discharge volume may be related to the success of medical treatment using the device.

[0057] l) Assembly force: This is used for assembling different parts of a housing or for fitting parts into the housing, such as connecting two parts to each other. Assembly forces can range from 10 N (Newtons) to 50 N. Component assembly is typically force-controlled. Samples exceeding these limits may be rejected. Therefore, a cutoff distribution appears when statistically analyzing these process control and / or monitoring parameters. 30 N could be an example. It may vary depending on the device / process step. Assembly force can be an indirect indicator of the overall device function. If the assembly force is too high or too low, it may indicate a problem and potential impairment of device function.

[0058] The above parameters can be determined and compared with nominal values, whether they have SI (International System of Units) units or not. In particular, these parameters are measurable, i.e., compared with SI units. Therefore, the requirements given in standards related to the SI system can be met.

[0059] Some force parameters may be related to mechanical energy storage devices within the medical device, such as mechanical springs. Examples of mechanical springs that can be used include helical springs, disc springs, and / or leaf springs. The springs can be tension springs or compression springs.

[0060] At least one limit of the tolerance range can be an upper limit compared to the upper limit of the specifications used for the production process of the device, which may also be valid for the sampled item. Comparisons can be made to evaluate the production process. If at least one tolerance limit is within the specification limits, the production process can be adjusted, for example. If at least one tolerance limit is outside the specification limits, additional measures are necessary, such as redesigning the device or using alternative production methods. This parameter can be a force parameter or another parameter. The technical limit can be the lower limit of a probability distribution function truncated on the left.

[0061] Alternatively, the at least one limit can be a lower limit compared to a lower limit of specifications used in the device manufacturing process. These specifications can also be valid for the sampling item. Comparisons can be made to evaluate the manufacturing process. Based on the evaluation results, the actions described above can be performed. The parameter can be the volume of drug expelled or the dosage accuracy. The technical limit can be the upper limit of a probability distribution function truncated on the right, for example, the maximum volume of the container containing the drug expelled or retained within the container.

[0062] The transformed probability content can be calculated using a specified probability content and a value determined by or equal to the area of ​​the truncated portion of the probability distribution function, i.e., the value of the cumulative distribution function of the probability distribution function considering the cutoff value. The truncated portion can be a range starting from negative infinity or starting at a corresponding value and ending at the cutoff value; that is, there will be a left truncation. Alternatively, i.e., for a right truncation, the truncated portion can be a range starting from the truncated value and ending at positive infinity or the corresponding value. Negative and positive infinity can be replaced with appropriately large values, for example, numbers in the range of 100 to 100 million. The area of ​​the truncated portion is a simple and descriptive feature that is easy to calculate. The actual probability content can be a linear transformation of the desired or specified probability content. The area of ​​the truncated portion can be subtracted from the specified probability content and normalized by dividing the difference by the area of ​​the remaining portion of the distribution.

[0063] The actual probability content can be calculated using a specified probability content, the cumulative distribution function of the probability distribution function, and a cutoff value. For lower or left truncation, the actual probability content can be calculated using the following formula:

[0064] PC act =(pc–F(ξ) L ,p)) / (1–F(ξ L ,p)),

[0065] Where pc specifies the probability content.

[0066] F(ξ L p) is the cumulative distribution function of the probability distribution function, starting from positive infinity or the corresponding value and ending at the lower cutoff value ξ. LFinish,

[0067] And p is a vector of characteristic or descriptive parameters of the probability distribution function, or a vector of approximate values ​​of the characteristic parameters. This is a relatively simple equation that is easy to compute, for example, using the low processing power and time of a digital electronic processor. This can be the first step e1) of providing the tolerance interval limit in step e).

[0068] Alternatively, the actual probability content of the upper or right truncation can be calculated using the following formula:

[0069] PC act =(pc–(1-F(ξ) R ,p)) / (F(ξ R ,p)),

[0070] Where pc specifies the probability content.

[0071] F(ξ R ,p) is the cumulative distribution function of the probability distribution function, and the upper cutoff value ξ is... R It begins and ends at infinity or the corresponding value, and p is a vector of characteristic or descriptive parameters of the probability distribution function, or a vector of approximate values ​​of the characteristic parameters. This equation is similar to a left-truncation equation, and is therefore easy to compute. This could be the first step e1) of providing the tolerance interval limit in step e).

[0072] In both cases, the characteristic or descriptive parameters can be the mean and / or standard deviation, especially when a normal distribution function is assumed and used in the calculation. Alternatively, other types of distribution functions can be assumed and used.

[0073] The transition probability can be calculated using a specified probability value and the actual probability value. The transition probability can also be calculated by calculating the specified probability value and the sum of the differences between the specified probability value and the actual probability value. For example, based on:

[0074] in It is the content of transformation probability.

[0075] pc specifies the probability content, and

[0076] PC act This refers to the actual probability content. This could be the second step (e2) that provides the tolerance interval limit in step e).

[0077] Therefore, the transformation of a specified probability content can be a linear function of the specified or desired probability content and a linear function of the area of ​​the truncated portion or the truncated portion. The area of ​​the truncated portion depends non-linearly on the truncation threshold, i.e., the cutoff value. The calculation of the transformed probability content can be derived from the following formulas (2) and (3).

[0078] The transformation probability can be used to calculate the tolerance limit factor, which is used to calculate at least one limit of the tolerance interval. At least one descriptive parameter of the probability distribution function can also be used to calculate at least one limit of the tolerance interval. When the probability distribution function is normally distributed, at least one limit of the tolerance interval can be calculated according to the following equation:

[0079]

[0080] Where UTBL is the upper limit of the tolerance interval.

[0081] LTBL is the lower limit of the tolerance range.

[0082] It is the average of the sample values.

[0083] It is the content of the transformation probability, and

[0084] s is the standard deviation of the sample values.

[0085] The descriptive parameters of the probability distribution function can be the same as the descriptive parameters of the cumulative probability distribution of the distribution function. This can be the third step (e3) that provides the tolerance interval limit in step e).

[0086] The article "An R package for Estimating Tolerance Intervals", Derek S. Young, Journal of Statistical Software, August 2010, Vol. 36, No. 5, pp. 1-39, especially Section 4.8, (univariate) normal tolerance intervals, mentions that in the case of a one-sided setting, there exists an exact solution for k, i.e., the one-sided tolerance interval. If transformation probability is used, the transformation tolerance limit factor can be calculated in the same way.

[0087]

[0088] Let n be a natural number representing the sample size, and t * n-1；1-alpha (delta) is the (1-α) quantile of a non-centralized t-distribution with d degrees of freedom (e.g., n-1) and a non-centralized parameter delta. P It is the P-quantile of the standard normal distribution. P is the proportion and the transformed probability content. This should be used for P. Sqrt() is the square root function. Alternatively, an exact solution can be computed via numerical integration. However, the numerical workload (i.e., processor power) can be significantly greater compared to using an analytical exact solution. This might be the first part of step e3). The computation of UTBL or LTBL could be the second part of step e3).

[0089] The device may be a medical device and, when calculating the at least one limit of the tolerance range, conforms to the requirements of ISO 11608-1 "Needle-based injection systems for medical use – Requirements and test methods – Part 1 Needle-based injection systems" 2014 or an earlier or later version thereof. Therefore, the current standard may be modified to include the proposed methods in the future. This allows for the production of devices with high-quality standards.

[0090] Tests can be performed to determine whether it is necessary to calculate the transition probabilities. The following test equations can be used: Less than or equal to 3*s,

[0091] in The average of the sample values.

[0092] s is the standard deviation of the sample values, and

[0093] || is the absolute value operator.

[0094] If the test equation is satisfied, the transformation probability content can be calculated and used to calculate at least one limit of the tolerance interval. Alternatively, if the test equation is not satisfied, only the probability content, rather than the transformation probability content, can be specified to calculate at least one limit of the tolerance interval.

[0095] On the other hand, a method for evaluating a production process is involved, including the same steps as a method for determining tolerance range limits. Furthermore, at least one limit of the tolerance range is compared with the limit of a specification range for a production item that has the same construction or the same construction design as the sampled item and is preferably manufactured according to the same specifications, e.g., product specifications or manufacturing specifications. If at least one limit of the tolerance range is located within or within the production specification range, the production process of the production item can begin, continue, or restart; alternatively, if the at least one limit exceeds the specification range, the production process of the production item may not start or may stop.

[0096] The methods used to evaluate the production process can be the methods used to determine tolerance range limits or the implementation schemes described above. Therefore, the same technical effects described above also apply to the evaluation methods for the production process.

[0097] The proposed method can be used in the industrial production process of preparing devices, preferably medical devices. The medical device can be a drug delivery device, more preferably an injection device, especially an auto-injector or a manually operated syringe. Prototypes or small-batch production can be carried out, for example, less than 100, 500, or 1000 units. If at least one limit of the tolerance range is within the specification range, the industrial production process can begin, i.e., for example, producing more than 1000 or 10000 units per year, or even more.

[0098] On the other hand, it relates to a computer program product having a computer-readable program code portion (instructions) that, when executed on a controller or processor, implements at least one, arbitrarily selected, or all of the method steps according to the above-described method. Therefore, the effective features, advantages, and technical effects of the proposed method and its implementations can also be effective for the computer program product. The computer program product can be used in many industries, such as medical device manufacturing, the automotive industry, the semiconductor industry, the solar energy industry, and the chemical industry. The computer program product can be a CD (optical disc), a data stream on the Internet, or data stored in RAM (random access memory), ROM (read-only memory, such as PROM (programmable ROM), EPROM (erasable PROM), or EEPROM (electronic EPROM)), hard disk, optical disc, solid-state device memory, or other digital storage devices.

[0099] Standard statistical software packages can be modified to simultaneously compute the transformation probability content and calculate at least one limit of the tolerance interval according to one or more proposed methods.

[0100] The final aspect concerns computing devices, such as computers, laptops, smartphones, tablets, etc., including:

[0101] - A processor configured to execute instructions (such as programs).

[0102] - A memory configured to store the instructions and data used or generated during the execution of the instructions.

[0103] - Preferably, a first data providing unit, such as an input device, is configured to input data that will be stored in memory and can be used during instruction execution, and

[0104] - Preferably, the second data providing unit, for example, is an output device configured to output data generated during instruction execution, and

[0105] - A third aspect of a computer program product or computer program product that calculates the transformation probability content based on the truncation of a probability distribution function used to calculate at least one limit of the tolerance interval.

[0106] Therefore, the effective features, advantages, and technical effects of the proposed method and its implementation scheme can also be effective for computing devices.

[0107] The first data providing unit can be an input file, keyboard, touchscreen, etc. The second data providing unit can be an output file, monitor, screen (preferably touchscreen), or other display device.

[0108] The manufacture and use of the presently preferred embodiments are discussed in detail below. However, it should be understood that this disclosure provides many applicable concepts that can be implemented in a wide variety of specific contexts. The specific embodiments discussed are merely illustrative of particular ways of making and using the disclosed concepts and do not limit the scope of the claims.

[0109] Furthermore, unless otherwise stated, the same reference numerals denote the same technical features. As used herein, the word "may" signifies both the possibility of doing so and the actual technical implementation. The concepts of this disclosure will be described in a more specific context, namely, the manufacturing process of a drug delivery device, with reference to the preferred embodiments described below. However, the disclosed concepts can also be applied to other situations and / or devices, such as those used in the manufacturing processes of automobiles, semiconductor products, or chemical products.

[0110] The features and technical advantages of embodiments of this disclosure have been outlined quite extensively above. Additional features and advantages of embodiments of this disclosure (e.g., the subject matter of dependent claims) will be described below. Those skilled in the art will understand that the disclosed concepts and specific embodiments can be readily used as the basis for modifying or designing other structures or processes for achieving the same or similar purposes as those specifically discussed herein. Those skilled in the art will also recognize that equivalent constructions do not depart from the spirit and scope of this disclosure as defined in the appended claims.

[0111] To gain a more complete understanding of the currently disclosed concepts and their advantages, the following description is now presented with reference to the accompanying drawings. The drawings are not drawn to scale. The following are shown in the drawings:

[0112] Figure 1 Drug delivery device,

[0113] Figure 2Test setup device for drug delivery devices

[0114] Figure 3 Empirical values ​​measured using a test setup device.

[0115] Figure 4 A computing device (computer) used to calculate the limits of the conversion tolerance zone.

[0116] Figure 5 Given the density function plot of the distribution, and

[0117] Figure 6 A plot of the density function of a given distribution with left truncation.

[0118] Figure 1 A drug delivery device 100 is shown that may include a container holding member 101. The drug delivery device 100 may include a main housing portion 102 that wholly or partially houses the container holding member 101 and includes other portions of the drug delivery device 100. Alternatively, the main housing portion 102 may be connected to the container holding member 101, but may not surround it, and may not even surround a portion of the container holding member 101; see [reference needed]. Figure 1 The dashed line in the middle.

[0119] The following can be arranged within the main housing portion 102:

[0120] - Piston rod 104, the piston rod being adapted to move a piston within the container holding member 301.

[0121] - A drive mechanism 106 for the piston rod 104. The drive mechanism 106 may include an energy storage element, such as a spring, which is manually or automatically loaded, for example, during assembly of the drug delivery device 100 or before each use.

[0122] - For example, an actuating element 108 at the proximal end P, which is used to initiate movement of the piston rod 104 into the container holding member 101, thereby using a drive mechanism 106. Alternatively, an autoinjector device actuated by axial movement of a movable needle cover may also be used.

[0123] - Cap 112, which may be attached to the main housing portion 102 or another portion of the drug delivery device 100. Cap 112 may be an outer cap that may include a smaller inner cap that may include a direct protective needle 110.

[0124] The drug delivery device 100 can be a single-use or multi-use device. For example, if the drug delivery device 100 is an auto-injection device, the actuating element 108 can be part of a triggering mechanism that is triggered from a remote location.

[0125] The medication can be dispensed from the container via needle 110 or via a nozzle that can be connected to and / or attached to the distal end D of the medication delivery device 100. Needle 110 can be replaced before each use or can be used multiple times.

[0126] The terms “drug” or “pharmaceutical preparation” are used synonymously herein and describe pharmaceutical preparations containing one or more active pharmaceutical ingredients or their pharmaceutically acceptable salts or solvates, and optionally a pharmaceutically acceptable carrier. In its broadest sense, the term active pharmaceutical ingredient (“API”) is a chemical structure that has a biological effect on humans or animals. In pharmacology, a drug or pharmaceutical preparation is used to treat, cure, prevent, or diagnose a disease or to otherwise enhance physical or mental health. A drug or pharmaceutical preparation may be used for a limited duration or periodically for chronic disorders.

[0127] As described below, a drug or pharmaceutical agent may include at least one API or combination thereof in various types of formulations for the treatment of one or more diseases. Examples of APIs may include small molecules (having a molecular weight of 500 Da or less); polypeptides, peptides, and proteins (e.g., hormones, growth factors, antibodies, antibody fragments, and enzymes); carbohydrates and polysaccharides; and nucleic acids, namely double-stranded or single-stranded DNA (including naked and cDNA), RNA, antisense nucleic acids such as antisense DNA and RNA, small interfering RNA (siRNA), ribozymes, genes, and oligonucleotides. Nucleic acids may be incorporated into molecular delivery systems such as vectors, plasmids, or liposomes. Mixtures of one or more drugs are also contemplated.

[0128] Drugs or pharmaceutical preparations can be contained in primary packaging or "drug containers" suitable for use with drug delivery devices. Drug containers can be, for example, cartridges, syringes, reservoirs, or other robust or flexible vessels configured to provide suitable chambers for storing (e.g., short-term or long-term storage) one or more drugs. For example, in some cases, the chambers can be designed to store the drug for at least one day (e.g., from 1 day to at least 30 days). In some cases, the chambers can be designed to store the drug for about one month to about two years. Storage can be carried out at room temperature (e.g., about 20°C) or refrigerated temperatures (e.g., from about -4°C to about 4°C). In some cases, drug containers can be or may include dual-chamber cartridges configured to separately store two or more components of the drug formulation to be administered (e.g., an API and a diluent, or two different drugs), one component in each chamber. In this case, the two chambers of the dual-chamber cartridge can be configured to allow mixing between the two or more components before and / or during administration to a human or animal. For example, the two chambers can be configured such that they are in fluid communication with each other (e.g., through a conduit between the two chambers), allowing the user to mix the two components before dispensing if needed. Alternatively or additionally, the two chambers can be configured to allow mixing during dispensing of the components into a human or animal body.

[0129] Drugs or agents contained in drug delivery devices as described herein can be used to treat and / or prevent many different types of medical disorders. Examples of disorders include, for example, diabetes or diabetes-related complications (such as diabetic retinopathy), thromboembolic disorders (such as deep vein or pulmonary thromboembolism). Further examples of disorders are acute coronary syndrome (ACS), angina pectoris, myocardial infarction, cancer, macular degeneration, inflammation, hay fever, atherosclerosis, and / or rheumatoid arthritis. Examples of APIs and drugs are those described in the following manuals: such as Rote Liste 2014 (e.g., but not limited to, main group 12 (antidiabetic drugs) or 86 (oncology drugs)) and Merck Index, 15th edition.

[0130] Examples of APIs used to treat and / or prevent type 1 or type 2 diabetes or complications associated with type 1 or type 2 diabetes include insulin (e.g., human insulin, or human insulin analogs or derivatives); glucagon-like peptide-1 (GLP-1), GLP-1 analogs or GLP-1 receptor agonists, or analogs or derivatives thereof; dipeptidyl peptidase-4 (DPP4) inhibitors, or pharmaceutically acceptable salts or solvates thereof; or any mixture thereof. As used herein, the terms “analyte” and “derivative” refer to a polypeptide having a molecular structure that is formally derived from the structure of a naturally occurring peptide (e.g., the structure of human insulin) by deletion and / or exchange of at least one amino acid residue present in a naturally occurring peptide and / or by addition of at least one amino acid residue. The added and / or exchanged amino acid residues may be codeable amino acid residues or other naturally occurring residues or purely synthetic amino acid residues. Insulin analogs are also referred to as “insulin receptor ligands”. Specifically, the term "derivative" refers to a polypeptide having a molecular structure that is formally derived from the structure of a naturally occurring peptide (e.g., human insulin), wherein one or more organic substituents (e.g., fatty acids) are bound to one or more amino acids. Optionally, one or more amino acids present in a naturally occurring peptide may have been deleted and / or substituted with other amino acids (including non-coding amino acids), or amino acids (including non-coding amino acids) may have been added to a naturally occurring peptide.

[0131] Examples of insulin analogs are Gly(A21), Arg(B31), Arg(B32) human insulin (glargine insulin); Lys(B3), Glu(B29) human insulin (glutamate insulin); Lys(B28), Pro(B29) human insulin (lispro insulin); Asp(B28) human insulin (aspart insulin); human insulin wherein the proline at position B28 is replaced by Asp, Lys, Leu, Val, or Ala and wherein Lys at position B29 can be replaced by Pro; Ala(B26) human insulin; Des(B28-B30) human insulin; Des(B27) human insulin and Des(B30) human insulin.

[0132] Examples of insulin derivatives are, for example, B29-N-myristoyl-des(B30) human insulin, Lys(B29)(N-tetradecanoyl)-des(B30) human insulin (detemir insulin, B29-N-palmitoyl-des(B30) human insulin; B29-N-myristoyl human insulin; B29-N-palmitoyl human insulin; B28-N-myristoyl LysB28ProB29 human insulin; B28-N-palmitoyl-LysB28ProB29 human insulin; B30-N-myristoyl-ThrB29LysB30 human insulin; B30-N-palmitoyl-ThrB29LysB30 human insulin; B29-N-(N-palmitoyl-γ-glutamyl)-des(B30) human insulin, B29-N-ω-carboxypentadecanoyl-γ-L-glutamyl-des(B30) human insulin (degludec insulin) ); B29-N-(N-lithochyl-γ-glutamyl)-des(B30) human insulin; B29-N-(ω-carboxyheptadecanoyl)-des(B30) human insulin and B29-N-(ω-carboxyheptadecanoyl) human insulin.

[0133] Examples of GLP-1, GLP-1 analogs, and GLP-1 receptor agonists include lixilatin. Exenatide (Exendin-4, Liraglutide, a 39-amino acid peptide produced by the salivary glands of the Gila monster. Semaglutide, Taspoglutide, Albiglutide Dulaglutide rExendin-4, CJC-1134-PC, PB-1023, TTP-054, Langnatide / HM-11260C, CM-3, GLP-1Eligen, ORMD-0901, NN-9924, NN-9926, NN-9927, Nodexen, Viador-GLP-1, CVX-096, ZYOG-1, ZYD-1, GSK-2374697, DA-3091, MAR-701, MAR709, ZP-2929, ZP-3022, TT-401, BHM-034, MOD-6030, CAM-2036, DA-15864, ARI-2651, ARI-2255, exenatide-XTEN, and glucagon-Xten.

[0134] Examples of oligonucleotides include, for instance, sodium mipronil. It is a cholesterol-reducing antisense agent used to treat familial hypercholesterolemia.

[0135] Examples of DPP4 inhibitors include vidagliptin, sitagliptin, denagliptin, saxagliptin, and berberine.

[0136] Examples of hormones include pituitary hormones or hypothalamic hormones or regulatory active peptides and their antagonists, such as gonadotropins (follicle-stimulating hormone, luteinizing hormone, human chorionic gonadotropin, fertility-stimulating hormone), growth hormone (Somatropine), desmopressin, terlipressin, gosorelin, triptorelin, leuprorelin, buserorelin, nafarelin, and goserelin.

[0137] Examples of polysaccharides include glucosaminoglycans, hyaluronic acid, heparin, low molecular weight heparin or ultra-low molecular weight heparin or their derivatives, or sulfated polysaccharides (e.g., polysulfated forms of the above polysaccharides), and / or their pharmaceutically acceptable salts. An example of a pharmaceutically acceptable salt of polysulfated low molecular weight heparin is enoxaparin sodium. An example of a hyaluronic acid derivative is Hylan GF 20. It is a type of sodium hyaluronate.

[0138] As used herein, the term "antibody" refers to an immunoglobulin molecule or its antigen-binding portion. Examples of antigen-binding portions of immunoglobulin molecules include F(ab) and F(ab')2 fragments that retain the ability to bind antigens. Antibodies can be polyclonal antibodies, monoclonal antibodies, recombinant antibodies, chimeric antibodies, deimmunized antibodies or humanized antibodies, fully human antibodies, non-human (e.g., mouse) antibodies, or single-chain antibodies. In some embodiments, antibodies have effector function and can fix complement. In some embodiments, antibodies have reduced or no ability to bind Fc receptors. For example, antibodies can be isotypes or subtypes, antibody fragments, or mutants that do not support binding to Fc receptors, for example, they have a mutagenic or missing Fc receptor-binding region. The term antibody also includes antigen-binding molecules based on tetravalent bispecific tandem immunoglobulins (TBTI) and / or antibody-like binding proteins with bivariate cross-binding domain orientation (CODV).

[0139] The term "fragment" or "antibody fragment" refers to a polypeptide (e.g., antibody heavy chain and / or light chain polypeptide) derived from an antibody polypeptide molecule that does not contain the full-length antibody polypeptide but still contains at least a portion of the full-length antibody polypeptide capable of binding an antigen. Antibody fragments may contain cleaved portions of the full-length antibody polypeptide, although the term is not limited to such cleaved fragments. Antibody fragments that can be used in this invention include, for example, Fab fragments, F(ab')2 fragments, scFv (single-chain Fv) fragments, linear antibodies, monospecific or multispecific antibody fragments (such as bispecific, trispecific, tetraspecific, and multispecific antibodies (e.g., double-chain, triple-chain, and quadruple-chain antibodies)), monovalent or multivalent antibody fragments (such as bivalent, trivalent, quadruvalent, and multivalent antibodies), microantibodies, chelated recombinant antibodies, tri- or bispecific antibodies, intracellular antibodies, nanobodies, small modular immunopharmaceuticals (SMIPs), binding domain immunoglobulin fusion proteins, camel-derived antibodies, and antibodies containing VHH. Further examples of antigen-binding antibody fragments are known in the art.

[0140] The term "complementarity-determining region" or "CDR" refers to a short polypeptide sequence within the variable region of both heavy and light chain polypeptides, primarily responsible for mediating specific antigen recognition. The term "framework region" refers to an amino acid sequence within the variable region of both heavy and light chain polypeptides; it is not a CDR sequence and is primarily responsible for maintaining the correct positioning of the CDR sequence to allow antigen binding. Although the framework region itself does not typically participate directly in antigen binding, as is known in the art, certain residues in the framework region of some antibodies can directly participate in antigen binding or can affect the ability of one or more amino acids in the CDR to interact with the antigen.

[0141] Examples of antibodies are anti-PCSK-9 mAbs (e.g., alirocumab), anti-IL-6 mAbs (e.g., sarilumab), and anti-IL-4 mAbs (e.g., dupilumab).

[0142] Pharmaceutically acceptable salts of any API described herein are also intended for use in drugs or pharmaceutical preparations in drug delivery devices. Pharmaceutically acceptable salts are, for example, acid addition salts and basic salts.

[0143] Those skilled in the art will understand that various components of the APIs, formulations, instruments, methods, systems, and embodiments described herein can be modified (added and / or removed) without departing from the full scope and spirit of the invention, which covers such modifications and any and all equivalents thereof.

[0144] Figure 2A test setup apparatus 200 is shown for testing at least one parameter of a drug delivery device, particularly a drug delivery device 100. The test setup apparatus 200 may include:

[0145] - Mounting device 201 that allows some components of test setup 200 to move vertically.

[0146] - Motor M, which generates torque for the movement of movable parts.

[0147] - Upper clamping device 202, which can clamp the distal or proximal end of the device under test.

[0148] - Lower clamping device 204, which can clamp the other end of the device being tested.

[0149] - Control device 206, which can control the motion generated by motor M, and

[0150] - Measurement reporting device 208, which is connected to, for example, a force sensor.

[0151] Other parts of the test setup 200 are not shown, such as optional scales, power supply units, etc.

[0152] The upper clamping device 202 and / or the lower clamping device 204 can move relative to each other to generate or apply a force to the device under test (DUT).

[0153] The test setup 200 can be used to measure forces associated with the drug delivery device 100 or other devices. In the following description, it is assumed that the test setup 200 is used to measure the cap attachment force of the cap 112 (see item d) in the introduction section of the specification. The drug delivery device 100 under test can be a device of device type U300max manufactured by the applicant of this application. However, other device types can also be tested.

[0154] The fully assembled drug delivery device 100 can be clamped in the test setup device 200. The cap 112 can be held by the lower clamping device 204. The proximal end P of the drug delivery device 100 can be held by the upper clamping device 202. However, it is also possible to hold the cap 112 in the upper clamping device 204 and the proximal end of the drug delivery device 100 in the lower clamping device 204.

[0155] Suppose that the cap adhesion of 19 drug delivery devices 100 was measured. These drug delivery devices 100 may be part of a prototype or small-batch production process. Alternatively, these devices may be drawn from a large-batch production process. An example of the test measurement is as follows: Figure 3 As shown.

[0156] Figure 3Empirical values ​​of cap adhesion force for 19 drug delivery devices 100, measured using test setup 200, are shown. The horizontal x-axis 300 is used to classify the sampled cap adhesion force x of cap 112 into classes with a width of, for example, 1 N (Newtons). Classes may exist with ranges from 0.5 N to 1.5 N (Class 1), 1.5 N to 2.5 N, 2.5 N to 3.5 N, 3.5 N to 4.5 N, etc. Sample values ​​fall within seven classes:

[0157] - Category 2 from 1.5N to 2.5N: 4 sample values,

[0158] - Category 3 from 2.5N to 3.5N: 2 sample values,

[0159] - Category 4 from 3.5N to 4.5N: 3 sample values,

[0160] - Category 5 from 4.5N to 5.5N: 5 sample values,

[0161] - Category 6 from 5.5N to 6.5N: 1 sample value,

[0162] - Category 7 from 6.5N to 7.5N: 2 sample values,

[0163] - Category 8 from 7.5N to 8.5N: 3 sample values.

[0164] The vertical y-axis 302 shows the number of samples within the corresponding category. Column 304 represents the frequency of the force values ​​of the cap attachment force. The average value is determined. In the example, the average It may have values ​​of approximately 6N. Furthermore, the standard deviation s is calculated for all 19 sample values ​​x.

[0165] Based on average Given the standard deviation s, the normal probability distribution function 306 (density function, PDF) can be calculated and mapped onto a histogram showing categories 2 to 8.

[0166] There exists a cutoff threshold of 308, also known as the technical limit TL, at which cap attachment may not occur. The value of threshold 308 is 0N.

[0167] Figure 3 The upper tolerance zone limit 310, UTBL, is shown without considering the truncation caused by the technical limit TL. Furthermore, the calculated lower tolerance zone limit 312, LTBL, is shown for the case where truncation due to the technical limit TL is not considered. The upper tolerance zone limit 310 (UTBL) and the lower tolerance zone limit 312 (LTBL) define the tolerance interval TI. The lower tolerance zone limit LTBL lies to the left of the technical limit 308, meaning its value is less than the technical limit 308.

[0168] Figure 3 The histogram shown can be a precise representation of the distribution of numerical data. It may be an estimate of the probability distribution of a continuous variable. The probability distribution function 306 can be truncated to take into account the technical limit 308 used to calculate the upper tolerance band 310. It is assumed that even considering the truncation, the lower tolerance band LTBL will not exceed the technical limit 308. Therefore, the tolerance interval TI is either a one-sided tolerance band or a tolerance interval TI with only the upper tolerance band UTBL. In particular, the tolerance interval TI is calculated as one-sided. However, when the distribution is in ξ... L When ξ is truncated, L The value can be interpreted as the lower limit of the tolerance zone.

[0169] In order to account for the technical limit 308 and calculate the upper limit of the tolerance zone UTBL, the cutoff value ξ must be determined. L This can be done graphically or through simple arithmetic calculations. In this example, another coordinate system with an x-axis of 300* can be used. The x-value of zero is where the maximum value of the probability distribution function 306 is located; that is, this corresponds to the mean value on the x-axis of 300. The location. ξ L The value corresponds to the average value.

[0170] Further optional steps may be necessary, such as determining the probability distribution function 306 and / or the value ξ before performing further calculations. L Normalization can be performed, or a standard statistical software package or a modified standard statistical software package can be used. The purpose of normalization is to make the area contained under the probability distribution function 306 have a value of 1. This may require compression or expansion in the x-direction and / or scaling in the y-direction. These steps can be performed automatically by executing the appropriate script in the statistical software package. Figure 6 The truncated probability distribution function 602 is shown. See below for reference. Figure 6 Describe the steps performed to calculate the upper limit of the conversion tolerance zone UTBL*, taking into account the technical limit 308.

[0171] Figure 4 A computing device 400 (computer) for calculating the limits of a transformation tolerance band is shown. The computing device 400 may include: - a processor (Pr) configured to execute instructions, particularly for performing the disclosed calculations.

[0172] - Memory (Mem), the memory being configured to store the instructions and data used or generated during the execution of the instructions.

[0173] - An optional input device (In), such as a keyboard, configured to input data that will be stored in memory (Mem), specifically input sample values ​​x.

[0174] - Optional output device (Out), such as a display device, configured to output data generated during instruction execution, and

[0175] - A computer program product that calculates the transformation probability content based on the truncation of a probability distribution function (306, 602) used to calculate at least one limit (UTBL, LTBL) of the tolerance interval (TI) (especially the upper limit UTBL* or lower limit LTBL* of the transformed tolerance band).

[0176] A connection / bus 410 may exist between the processor Pr and the memory Mem. Other units of the computing unit 400 are not shown but are known to those skilled in the art, such as power supply units, optional internet connectivity, etc. Alternatively, a server solution may be used, utilizing computing power and / or storage space available on the internet or corporate intranet provided by other service providers.

[0177] Figure 5 A graph of the density function 502 for a given distribution is shown, with the upper tolerance band UTBL and the upper specification band USL. The x-axis 500 shows the values ​​of the sampled values ​​x. The y-axis is not shown, but is used to show the frequency of each sampled value x. Figure 5 The example shown uses the normal probability distribution function 502 (density function), also known as a bell curve or Gaussian curve. The area under the probability distribution function 502 is 1. This is illustrated by referring to the cumulative distribution function F(x,p), CDF, of the normal probability distribution function 502 (PDF). Figure 5 The upper tolerance band 504, UTBL, is shown. When calculating the upper tolerance band 504, UTBL, using a confidence level 1–α, probability content or proportion pc, and a vector p containing descriptive parameters of the normal probability distribution function 502, truncation is not considered. This calculation is known to those skilled in the art. The upper tolerance band 504 is lower than the upper specification band 506, USL. Therefore, a decision may be made to begin the production process of the drug delivery device 100. However, this decision may be erroneous because the technical limit 308 is not properly considered.

[0178] Figure 6 A graph of the probability distribution function (density function) 602 (e.g., 306) for a given distribution is shown, where ξ L Left truncation. If the probability content of the transformation is used in the calculation. The upper tolerance band UTBL is then converted to the upper tolerance band UTBL* of the transformation. In the example, the probability content of the transformation... It is still less than the upper limit of the specification, USL. However, compared to Figure 5Compared to the original tolerance zone upper limit UTBL, the distance between the converted tolerance zone upper limit UTBL* and the specification upper limit decreases because the converted tolerance zone upper limit UTBL* shifts to the right (see delta 612). This is due to the consideration of the left truncation ξ. L . and based on Figure 5 The decisions made by the drug delivery device 100 are more reliable than those made by the manufacturer.

[0179] Figure 6 The x-axis (600) is shown for the sample value x. The y-axis is not shown, but it relates to the frequency of each sample value. The normal probability distribution function (602, density function), PDF, approximates... Figure 3 Histogram of samples of the drug delivery device 100 or other device mentioned in the description. The cutoff threshold 604 corresponds to the technology limit TL, for example, to zero; see also [link to other documentation]. Figure 3 The area of ​​606 corresponds to the cutoff value ξ. L Or cutoff value ξ L The value of the cumulative distribution function F at point 606. If we subtract the cutoff area 606, the area 608 corresponds to the remainder of the cumulative distribution function F, that is, the value 1-F(ξ). L ,p).

[0180] Line 610 illustrates the case where the probability distribution function 602 is truncated by the technical limit 308, i.e., at the cutoff value ξ. L The location is the upper limit of the tolerance zone of 610 (UTBL*) for the transformation. The delta value of 612 is shown to be particularly large to make the displacement visible. Note that the displacement may also occur in the other direction, i.e., UTBL* may be less than UTBL, depending on the cutoff value ξ. L The value. Line 614 shows an example of the specification upper limit USL.

[0181] The upper limit of the transformed tolerance band is 610 (UTBL*) based on confidence level 1-α and the transformed probability content. The descriptive parameter vector of probability density function 602 is calculated. The transformed probability content... The calculation is performed according to formulas (1) and (2), which will also be described below. The intermediate step in calculating the upper limit of the converted tolerance zone 610 (UTBL*) is to calculate the converted tolerance limit factor according to formula (5) described below.

[0182] The upper limit of the conversion tolerance zone UTBL* can be calculated using the following formula:

[0183]

[0184] in, s is the mean of the sample values ​​x, and s is the standard deviation of the sample values. The calculation of the mean and standard deviation can be learned from basic statistics textbooks.

[0185] If the probability distribution function has a right truncation, then the corresponding calculations can be performed. In this case, the lower limit of the conversion tolerance LTBL* can be calculated using the following formula:

[0186]

[0187] in, s is the mean of the sample values ​​x, and s is the standard deviation of the sample values. The calculation of the mean and standard deviation can be learned from basic statistics textbooks.

[0188] In other words, besides the above Figures 1 to 6 Beyond its description, in statistics, a truncated distribution is a conditional distribution that arises from restricting the domain of certain other probability distributions. In practical statistics, a truncated distribution occurs when, for some reason, the range of values ​​is restricted to values ​​above or below a given threshold or within a specified range. For example, if the force required to remove the cap 112 of device 100 is measured, this distribution will be truncated because the force can only be positive.

[0189] Therefore, several device-related characteristics must be evaluated, which may be truncated. For example, the discharge volume of pen device 100 is always non-negative. Therefore, the dose accuracy distribution is within the threshold ξ. L The value at 0 is left-truncated. In practice, feature truncation is usually negligible because the actual values ​​considered are far enough from the corresponding truncation threshold that their impact on the derived statistics is insignificant. However, in some cases, it can have a significant impact.

[0190] It can be stipulated that, in verification testing, the statistical tolerance interval TI should be calculated according to ISO 16269-6 (Statistical Interpretation of Data – Part 6: Determination of Statistical Tolerance Intervals, 2014(E), or an earlier or later version thereof) and compared with the specification range S, where truncation due to physical limitations can be considered a limitation of the specification range. The acceptance criterion is that the tolerance interval TI is part of or identical to the specification range S. A graph of the density function of the upper tolerance band UTBL and the upper specification limit USL for a normal distribution or another given distribution is shown in [the figure]. Figure 5 In this case, we have UTBL≤USL, meaning the tolerance interval TI is a part of or the same as the specification limit S. Therefore, it meets the acceptance criteria.

[0191] In some cases, a tolerance zone limit may exist that exceeds the corresponding physical limit, even though it is known that this physical limit will never be exceeded in practice. In such cases, the violation can be considered practically nonexistent. However, the other side of the tolerance zone may be affected, and this effect can be appropriately handled. It is important to note that this situation differs from what is known as censoring, where values ​​below or above a threshold cannot be observed due to technical or physical limitations or characteristics of the measurement system, even though these values ​​exist both in practice and theory. The suggested approach should not be applied to "reviews" as there may be other computational applications.

[0192] To facilitate understanding of the invention, the following formulas are provided. However, the invention should not be limited to this theory or any other theory. The values ​​of a given feature can be statistically distributed by a given density function f or probability distribution function PDF, which can be a mapping of real numbers (e.g., sample values ​​x) to an interval of 0 to 1, and in particular, also includes the limits of the interval. This can be expressed as: f(.;p):R->[0;1], where the elements of the parameter vector p are R n , where n is the dimension of the parameter space. Typically, for a distribution used for medical device 100 or other types of devices, n = 2. For example, in the case of a normal distribution, the parameter vector p is p = [μ; σ^2]. T , where “μ” is the expected value (mean), and σ^2, which is the square of the standard deviation σ, is the variance σ^2. T denotes the transpose of the vector.

[0193] The corresponding distribution function or cumulative distribution function (CDF) is:

[0194] f(·;p):R->[0;1],

[0195] The "·" symbol is a placeholder.

[0196] at the same time:

[0197]

[0198] Where “a” is a special value for the device characteristic.

[0199] Furthermore, it may be effective:

[0200] ∫ R f(·,p)=1, if in R n Integrating over the entire defined domain, i.e., the n-dimensional space of real numbers, where n is a natural number greater than or equal to 1.

[0201] According to standard ISO 16269-6:2014(E), the introduction, the first sentence:

[0202] "A statistical tolerance interval is a sample-based estimated interval that can be asserted with a confidence level of 1-α (e.g., 0.95) to include at least a specific proportion p of the items in the population." This definition of the statistical tolerance interval TI is also used in this application. In this application, the specified portion p is named the probability content pc.

[0203] The standard ISO 16269-6:2014(E) discloses the possibility of calculating the tolerance interval TI, specifically the lower and upper limits of the tolerance interval. However, truncation is not considered.

[0204] Without losing generality, the case of left truncation is considered below, see [link to relevant section]. Figure 6 The figure shows the density plot for a given distribution function 602, where ξ L The left truncation is present. This means that ξ... L With R's ξ L The elements exist together, therefore F(ξ) L The probability content of p) (this is the truncated left-hand side) does not actually exist. The upper tolerance band UTBL is based on the transformed probability content. Convert to the upper limit of the conversion tolerance zone UTBL*. Additionally, the upper limit of the specification USL is also shown.

[0205] This means that there exists ξ in R. L Elements such that F(ξ) L The probability content pc of ξ is p), i.e., the truncated ξ. L The part on the left, 606, does not actually exist, because, as mentioned above, the variable x under study may have a technical limit, 308.

[0206] The upper tolerance band (UTBL) indicates that a certain probability content pc lies within the corresponding tolerance interval TI. However, in the actual unrelated probability content pc, part F(ξ) L In critical cases, the tolerance band upper limit (UTBL) may require some transformation to reflect the cutoff of the cumulative density function (CDF).

[0207] As mentioned above, it may be necessary to convert the required probability content pc. ),like Figure 6 As shown, where T L It is a transformation function that is effective for left truncation.

[0208] In the first step, we can calculate the actual probability content pc. act , or the actual probability content pc covered by the tolerance interval TI act For example, due to the truncation of ξ L This can be corrected by subtracting the corresponding part 606. More specifically, the actual probability content pc actIt can be calculated through a linear transformation of the expected probability content. The truncated portion 606 of the distribution function 602 (see...) Figure 6 The left part of the distribution (in the original text) can be subtracted from the expected probability content pc. This can be done by dividing by the remaining part of the distribution, 608 (see [reference]). Figure 6 Normalize the right part of the text.

[0209] PC act =(pc–F(ξ) L ,p)) / (1–F(ξ L Formula (1)

[0210] However, other formulas can also be used to determine the actual probability content pc. act .

[0211] In the second step, specify / required probability content pc and actual probability content pc. act The difference between them can be added to the desired probability content pc to obtain the transformation probability content.

[0212]

[0213] However, other formulas can also be used as transformation function T. L (pc). Using formula (1), the transformation function T L (PC) = PC + (PC – PC) act This applies to the following situations:

[0214]

[0215] =(2-1 / (1–F(ξ)) L ,p))*pc+F(ξ L ,p) / (1–F(ξ L Formula (2)

[0216] Therefore, the transformation of the expected probability content pc can be a linear function of the expected probability content pc and the truncated portion 606 of distribution 602. The area of ​​the truncated portion 606 can depend non-linearly on the truncation threshold ξ. L .

[0217] In R n ξ R When right-truncation occurs at the element, the corresponding conversion of the probability content is:

[0218]

[0219] =T R (pc)=pc(2-1 / (1–F(ξ R ,p))+(1–F(ξR ,p)) / F(ξ R Formula (3)

[0220] The derivation of the right-hand truncation formula involves further terminology conversion, directly similar to the left-hand truncation case. Please refer to section pc in the introduction of the instruction manual. act And the right truncation formula.

[0221] Example: Suppose the force characteristic of a pen device 100 is tested, and the distribution of the sample values ​​x is known to be normally distributed. For this characteristic, a USL of 40 N (Newtons) is defined. It is well known that, due to physical properties, the force cannot be less than ξ. L =0N. The acceptance criterion for the experiment can be that, at a confidence level of 1–α = 95%, at least pc = 97.5% probability content is within the specification limits. The experiment was conducted on 20 samples, and the average value was... The estimated standard deviation is s = 5.97N. The tolerance zone limit is calculated using x + / - k*s, without considering truncation, over a bilateral tolerance interval. Here, k = 3.154 is the tolerance limit factor under given conditions. This results in a tolerance interval TI = [-9.70; 27.96]N. In this case, the lower tolerance zone limit LTBL is below the truncation threshold ξ. L .

[0222] To properly handle the truncation problem, the method derived above can be used. The required probability content can be converted using formula (2):

[0223]

[0224] in

[0225] (1 / (5.97*sqrt(2*PI)*integral from–infinity(∞)to 0over exp(-1 / 2*((x-9.13) / (5.97))^2dx)=0.06

[0226] Where PI = 3, 24, ..., sqrt() is the square root function, and exp is the exponential function.

[0227] And further use formula (2):

[0228]

[0229] In the third step, for Determine the corresponding tolerance limit factor Right now Standard statistical software packages can be used to calculate However, the probability content of the transformation must be used. Instead of the specified probability content pc recommended by standard statistical software packages.

[0230] In step four, the calculated tolerance range is I = [0; 26.12]N. The upper limit is also lower than the specification upper limit USL of 40N. Therefore, it meets the acceptance criteria.

[0231] The calculation can be performed using commercially available software packages. Software R can be used, especially the tolerance interval TI package; see "An R package for Estimating Tolerance Intervals", Derek S. Young, Journal of Statistical Software, August 2010, Vol. 36, No. 5, pp. 1-39, especially Section 4.8, (univariate) normal tolerance intervals. The relevant function can be:

[0232] K.factor(n,f=NULL,alpha=0.05,P=0.99,side=1,method=c("HE","HE2","WBE","ELL","KM","EXACT","OCT"),m=50), however It can replace pc for the value P of the function. It can be set using a one-sided setting, i.e., side = 1.

[0233] Derek S. Young's article mentions that, in the case of a one-sided setting, k has an exact solution:

[0234] k = 1 / sqrt(n)t * n-1；1-alpha (sqrt(n)z P ) Formula (5)

[0235] Let n be the sample size, t * n-1；1-alpha (delta) is the (1-α) quantile of a non-centralized t-distribution with d degrees of freedom (e.g., n-1) and a non-centralized parameter delta. PIt is the P-quantile of the standard normal distribution. Alternatively, the exact solution can be computed through numerical integration. However, the numerical workload (i.e., processor power) can be much greater compared to using the analytical exact solution. For example, the importance of truncation can be assessed according to Barbara Bredner's rules in the following articles: "Prozessfaehigkeit bewerten, Kennzahlen fuer nomalverteilte und nicht-normalverteilte Merrkmale", June 11, 2014, http: / / www.drsteuer.de / vorlagen / Prozessfaehigkeit_bewerten_Nov_14.pdf (accessed May 16, 2019) and "Prozessfaehigkeit bei technisch begrenzten Merkmalen, Faehigkeitskennzahlen und Berechnungsmethoden", January 24, 2014, URL: https: / / www.drsteuer.de / vorlagen / Prozessfaehigkeit_bei_technisch_begrenzten_Merkmalen_Jan_14.pdf. It is recommended that... When the time is less than or equal to 3 seconds, the cutoff is relevant, where It is the sample mean, and ξ is the cutoff threshold. L or ξ R “s” is the estimated standard deviation of the sample.

[0236] Further steps could include an analysis of the processing power of the truncated distribution, as described by Barbara Bredner in the following article: “Prozessfaehigkeit bei technisch begrenzten Merkmalen, Faehigkeitskennzahlen und Berechnungsmethoden”, January 15, 2015, URL: https: / / www.bb-sbl.de / wp-content / uploads / 2018 / 08 / BB-SBL-Prozessf%C3%A4higkeit-bewerten-V13-2015-01-15.pdf (accessed May 20, 2019).

[0237] In summary, the conversion of the required probability content for truncated distributions is disclosed. In the case of left truncation, the probability content should be converted according to formula (2) or a similar / corresponding formula. In the case of right truncation, the probability content should be converted according to formula (3) or a similar / corresponding formula.

[0238] A given technical instruction can also be converted to a distribution other than the normal distribution, such as the log-normal distribution, Weibull distribution, Gumbel distribution, Frechetz distribution function, etc.

[0239] Although embodiments of this disclosure and their advantages have been described in detail, it should be understood that various changes, substitutions, and alterations may be made therein without departing from the spirit and scope of this disclosure as defined by the appended claims. For example, those skilled in the art will readily understand that many of the features, functions, processes, and methods described herein can be varied while remaining within the scope of this disclosure. Furthermore, the scope of this application is not intended to be limited to specific embodiments of the systems, processes, manufactures, methods, or steps described herein. As will be readily understood by those skilled in the art from the disclosure of this disclosure, existing or later-developed systems, processes, manufactures, methods, or steps that perform substantially the same functions or achieve substantially the same results as the corresponding embodiments described herein can be utilized based on this disclosure. Therefore, the appended claims are intended to include such systems, processes, methods, or steps within their scope. The embodiments mentioned in the first part of the description can be combined with each other. Figures 1 to 6 The described implementation schemes can also be combined with each other. Furthermore, the implementation schemes mentioned in the first part of the description can be combined with those involving... Figures 1 to 6 The second part of the description is a combination of examples.

[0240] List of reference numerals

[0241] 100 Drug Delivery Devices

[0242] 101 Container holding member

[0243] 102 Main Casing Section

[0244] 104 Piston Rod

[0245] 106 Drive Mechanism

[0246] 108 Actuating Components

[0247] 110 needles

[0248] 112 hats

[0249] 200 Test setup device

[0250] 201 Installation equipment

[0251] 202 Upper clamping device

[0252] 204 Lower clamping device

[0253] 206 Control device

[0254] 208 Measurement Reporting Device

[0255] 300 x-axis

[0256] 302 y-axis

[0257] 304 represents a column representing a category of force values.

[0258] 306 Normal probability distribution function

[0259] 308 Cutoff Threshold

[0260] 310 Tolerance with upper limit UTBL

[0261] 312 Calculated lower limit of tolerance zone LTBL

[0262] 400 computing devices

[0263] 410 connection

[0264] Pr processor

[0265] Mem memory

[0266] In input device

[0267] Output device

[0268] 500 x-axis

[0269] 502 Normal probability distribution function

[0270] 504 Stainless Steel with upper tolerance limit UTBL

[0271] 506 Specification Upper Limit USL

[0272] 600 x-axis

[0273] 602 Normal probability distribution function

[0274] 604 Cutoff Threshold

[0275] 606. Area corresponding to the cumulative distribution function F value

[0276] 608 is the area corresponding to the remainder of the cumulative distribution function F.

[0277] 610 Tolerance with upper limit UTBL

[0278] 612 delta value

[0279] 614 Specification Limit USL

[0280] TL Technology Limits

[0281] CDF cumulative distribution function

[0282] PDF probability distribution function

[0283] x The measured variable

[0284] A set of values ​​for X x

[0285] Standard deviation of s, σ X

[0286] σ^2 variance (lowercase Greek letter Sigma)

[0287] μ Expected value (lowercase Greek letter My)

[0288] S Specification Range

[0289] TI tolerance range

[0290] DE device evaluation

[0291] USL Specification Upper Limit

[0292] UTBL tolerance zone upper limit

[0293] UTBL* Upper limit of tolerance band conversion

[0294] LSL specification lower limit

[0295] LTBL tolerance band upper limit

[0296] 1-α confidence level (1 – lowercase Greek letter Alpha)

[0297] F(·p) is the cumulative distribution function with parameter vector p.

[0298] f(·p) is the probability distribution function (density function) with parameter vector p.

[0299] k tolerance limit factor

[0300] Conversion tolerance limit factor

[0301] PC probability content

[0302] Conversion probability content

[0303] R real numbers

[0304] x T Transpose of vector x

[0305] The average value of x

[0306] ξ General cutoff threshold (lowercase Greek letter Ξ)

[0307] ξ L Left truncation threshold

[0308] ξ R Right truncation threshold

[0309] Newton (N) is the unit of force.

Claims

1. A method for determining the limits of a tolerance interval TI, comprising the following steps: -a) Provides multiple sample values ​​x, wherein the sample values ​​x fluctuate and define a sample value distribution (306, 602), wherein the sample values ​​x are technical parameter values ​​related to the sampling items of the samples, wherein the sampling items are parts of the drug delivery device (100), components for the drug delivery device (100), or the drug delivery device (100), wherein the sampling items have the same construction, and wherein the technical parameters are limited by at least one technical limit value. -b) Select the probability distribution function (306, 602) based on the technical parameters and / or the sample value x. -c) Use the technical limit value (308) to determine the cutoff value ξ of the probability distribution function (306, 602). L ξ R , -d) Specifies the probability content pc of the tolerance interval TI, and -e) Based on conversion probability content Provide the limits UTBL and LTBL of the tolerance interval TI for the technical parameters, wherein the transformation probability content Based on the cutoff value ξ L ξ R The probability content pc of the tolerance interval TI specified in step d).

2. The method of claim 1, wherein the probability content pc of the tolerance interval TI specified in step d) refers to the proportion or percentage of the total number of production items, wherein the production item is a part of a drug delivery device, a component for a drug delivery device, or a drug delivery device (100) constructed identically to the sampled item. The total amount of the production items mentioned therein has been produced or will be produced, and is at least 10 times the size of the sample from which the sample value x is derived.

3. The method of claim 1, wherein the probability content pc of the tolerance interval TI specified in step d) refers to the proportion or percentage of the total number of production items, wherein the production item is a part of a drug delivery device, a component for a drug delivery device, or a drug delivery device (100) constructed identically to the sampled item. The total amount of the production items mentioned therein has been produced or will be produced, and is at least 100 times the size of the sample from which the sample value x is derived.

4. The method of claim 1, wherein the probability content pc of the tolerance interval TI specified in step d) refers to the proportion or percentage of the total number of production items, wherein the production item is a part of a drug delivery device, a component for a drug delivery device, or a drug delivery device (100) constructed identically to the sampled item. The total amount of the production items mentioned therein has been produced or will be produced, and is at least 1000 times the size of the sample from which the sample value x is derived.

5. The method of claim 1, wherein at least one descriptive parameter of the sample value x s or at least two descriptive parameters s is calculated. Wherein at least one descriptive parameter is used s, or wherein at least two of the descriptive parameters are used. s, to calculate the cutoff value ξ of the probability distribution function (306, 602). L ξ R .

6. The method according to claim 1, wherein the technical parameter is one of the following parameters of the drug delivery device (100): Dosage accuracy, selector torque, dispensing force, cap (112) attachment force, cap (112) removal force, needle cover removal force, injection time, activation force, needle cap blocking distance, needle extension amount, discharge volume, assembly force.

7. The method according to claim 1, wherein a) The at least one limit of the tolerance range is the upper limit UTBL, which is compared with the upper limit of the sampled item's specification USL to evaluate the production process, or b) wherein the at least one limit is a lower limit LTBL, which is compared with the specification lower limit LSL of the sampled item to evaluate the production process.

8. The method according to claim 1, wherein the conversion probability content The probability content pc and value of the tolerance interval TI specified in step d) are calculated, and the value is determined by or equal to the area of ​​the cutoff portion (606) of the probability distribution function (306, 602), or the value is calculated by taking into account the cutoff value ξ. L ξ R The value of the cumulative distribution function F of the probability distribution function (306, 602) is determined or equal to the value of the cumulative distribution function, wherein the truncated portion (606) starts from negative infinity or the corresponding value and ends at the cutoff value ξ. L Within the range of the end, or from the cutoff value ξ R The range that begins with positive infinity or the corresponding value.

9. The method of claim 8, wherein the actual probability content pc act The probability content pc of the tolerance interval TI specified in step d), the cumulative distribution function F of the probability distribution function (306, 602), and the cutoff value (308) are used to calculate the result.

10. The method of claim 9, wherein the conversion probability content Use the probability content pc of the tolerance interval TI specified in step d) and the actual probability content pc act The calculation yielded the result.

11. The method of claim 10, wherein the conversion probability content Used to calculate tolerance limit factor The tolerance limit factor is used by employing at least one descriptive parameter of the probability distribution function (306, 602). s is used to calculate at least one limit UTBL, LTBL of the tolerance interval TI.

12. The method of claim 11, wherein the tolerance limit factor is calculated. For unilateral settings, based on: Where n is a natural number representing the sample size, t * n-1；1-alpha (δ) is the (1-α) quantile of a non-centralized t-distribution with n-1 degrees of freedom and a non-centralization parameter δ, z P It is the P-quantile of the standard normal distribution. The conversion probability content For P, Where δ is sqrt(n)z P ,and sqrt() is the square root function.

13. The method according to any one of claims 1-10, wherein the device (100) is a medical device, and wherein the at least one limit UTBL, LTBL of the tolerance interval TI is calculated in accordance with the requirements of ISO 11608-1 "Medical needle injection systems - Requirements and test methods - Part 1 Needle injection systems" 2014 or earlier or later versions thereof.

14. The method according to any one of claims 1-10, wherein the at least one limit of the tolerance interval TI is used for the manufacture of the drug delivery device (100) or related parts or components.

15. The method of claim 14, wherein the at least one limit of the tolerance interval TI is used to adjust the production process.

16. The method of claim 10, wherein the conversion probability content By calculating the sum of the probability content pc of the tolerance interval TI specified in step d), and the sum of the probability content pc of the tolerance interval TI specified in step d), and the actual probability content pc act It is calculated by summing the differences between them.

17. A method for determining the limit of a tolerance interval TI, wherein a judgment is performed on whether it is necessary to calculate the transformation probability content. The test, If the test equation is satisfied, the method according to any one of claims 1-10 is performed; and if the test equation is not satisfied, the probability content pc of the tolerance interval TI specified in step d) is used to calculate the at least one limit UTBL, LTBL of the tolerance interval TI, but not to calculate the transformation probability content.

18. A method for evaluating a production process, comprising the step of any one of claims 1-10, wherein the at least one limit UTBL, LTBL of the tolerance interval TI is compared with the limit USL, LSL of a specification interval for a production item having the same construction as the sampled item.

19. A computer program product having a computer-readable program code portion, which, when executed on a controller or processor Pr, implements at least one, any selected plurality or all of the method steps according to any one of claims 1-10.

20. A computing device, comprising: The processor Pr is configured to execute instructions. The memory Mem is configured to store the instructions and data used or generated during the execution of the instructions. The computer program product according to claim 19 or the calculation of the transformation probability content based on the truncation of the probability distribution function (306, 602) of at least one limit UTBL, LTBL for calculating the tolerance interval TI. Computer program products.

Images